yellowshark

Hi Illuminarti/George If you have a .4 nozzle and define it in Cura (for example) as .38 or .42 Does that have any effect on the statements above?

Joatrash , I think it is important to understand 2 points about the technology you are using, i.e. FDM You need to closely assess each model you wish to print and decide on the best orientation of the model on the print bed i.e. that which best suits the way the printer works, its strengths and weaknesses (when I say printer I mean the FDM process method). It may even be easier/better to split the model and then glue together. Unlike the other mainstream 3D print methods, FDM does not lend itself well to supports, unless you have dual extruders; even then the current capability of the software slicers is pretty damn awful in managing the use of dual extruders – which I am sure will change in the coming year or two as more manufacturers develop dual extrusion from the ground up. Yes you can get away with single extrusion supports IF you are supporting internal unseen parts of the model, but not external faces – unless you are prepared to do a lot of post print finishing. 2. Good designers design not just to meet the client’s requirements but also to meet the needs of the manufacturing process employed. It is quite probable that if you download a model from the internet it will not have been designed for FDM, it may well have been designed for CNC machining or injection moulding. I have just finished modifying a part for a client which comes from an assembly which was injection moulded and designed as such. During the modification design process I suggested a couple of minor changes that would make the job using FDM so much easier and the client was agreeable. As a result not only did he get the part in 24 hours he got it a lot cheaper. If you do design your own 3D models then you need to ensure you really understand FDM and as much as you can take that into account. Sorry, I am not suggesting you do not, I am just trying to articulate the general point. As a piece of hardware the Ultimaker is capable of producing excellent results but it is all the other aspects that you and we have to take responsibility for; it applies to any FDM printer.

Had that very very very occasionally, including today but as Kenneth says it is normally cleared after a few layers. Never spent any time trying to understand it and always just assumed it was "dirt" on the bed and take it as a sign to clean the bed

Ditto for me, the same as Robert and Illuminarti. I tend to buy 40 or 50m lengths from Faberdashery and the only other thing I do on first use is to recoil the filament in the same size loops as Fab to just make sure there are no tangles. I have now got to the point where I prefer to use these "loose" loops from the floor rather that the spools, as they act consistently all the way through unlike spools, which can be problematic once you get to the last third or so of the filament

Personally I would say there is no need to exceed 60c for PLA even for pieces taking the entire bed area; I have never intentionally gone above 60c. I have had the effect on some pieces and overcome it by reducing the area of the bottom 0.3mm (i.e. layer 1) of the model by 0.1mm all-round. Effectiveness of course probably does depend on the geometry but e.g. I always design a cylinder with 0.1mm taken off the diameter if I remember.

It will be the reduction in layer time as you approach the "point", almost certainly. IE subsequent layers are laid down before the previous one has dried and they melt together. It depends on the size and the geometry but slowing down the print speed as you approach the issue area can achieve the desired result. If not you will need to print concurrent items to increase the layer time at the end; pia I know

LOL Schafe you sound like a dangerous person, perhaps you should live in NW Washington

Dutch orange is my favourite filament, it lays down so well. If I have a problem with a filament I will load Dutch orange to check whether the problem is with the filament or the printer. I would run Colorfabb with .06 at 40m/s at no more than 210, quite possibly 200-205 - no idea about a UM2 though.

Thanks for your help guys, much appreciated. Joergen, I saw the good ABS article but I did not see anything on nylon. ?? Jeremie, yes that is a good article; I must read again more slowly and absorb it all . Interestingly though the Taulman website does talk about 645 being a different nylon, so I do not know if everything in the article is relevant. I say this because… Early this evening(before seeing your posts) I decided to proceed - as I need to have a sample ready for a customer on Friday. With my earlier failure with their T-glass I expected to spend hours and hours and hours. Well heck, it worked first time J!!! The reel was still in its original packaging and I know I should still have dried it but it was about 17:00 hours and I knew that any suggestion that I needed the oven for a couple of hours would not have been received well by my wife! I printed the standard test cube, setup as… Temp 235c Speed 20 m/s Bed temperature – none Layers .2 (.3 for first) Fans 100% from layer 2 20 line brim Glass plate with hairspray (my standard PLA base) The first amazing result was that the brim just stuck immediately – the smile on my face was immense! And off it went. After about 1.5mm I got some under-extrusion. Hmnn I thought, so I increased the pressure slightly on the filament driver and that solved it. At the end I thought I could see a few spots of under-extrusion so I did it again with the temp at 245, which is actually Taulman’s recommended temp. Great result, no warping at the corners, pretty good corners being vertically straight and reasonably sharp and no hint of rhino horns. Of course with all the steam that was pouring out during the process the finish is frosted. I then tried to run at 30 m/s but got under-extrusion, having finally pushed the temp. to 255. I figured it could well be due to the moisture content so I stopped. (of course it could be due to your point Jeremie about the thickness of the filament) I went back to 20m/s and printed one of my pieces which is flat, about 1mm high but with a much larger surface area, about 8cm * 5cm; I got another good result without a hint of the corners warping. So I am really pleased so far and the final piece demonstrated the flexibility that I think my customer wants. I have cut 10mtrs off the reel and that is going in the oven at 08:00 tomorrow for a couple of hours and then I continue; to try and push a higher speed and try a more complex geometry.

Thanks for the help guys, much appreciated. I thought about this some more this morning and recalled that I had seen these gappy brims when I was testing/analysing what was the best Z-offset distance for me. So I checked and saw that the start g-code I had used in Cura was very marginally different to what I had been using (the method of definition between the two is different). So I changed the distance an reran and I am please to say it was successful Well I say successful, one of the two pointy bits had a nice rhino horn and it is clear that a few layers from the end the nozzle tip caught on this. The piece was so well stuck to the bed that it did not move, but the glass bed did!! Apart from that I am very pleased with the end result.

Hi, having been using Slic3r I thought it was about time to try CURA again. So yesterday I downloaded the latest version. My first 4 attempts failed so I switched to Slic3r and it worked 1st time. As you can see the part is quite small On each occasion the part came away from the bed - something I just really do not suffer from. On closer inspection I saw that of the brims had gaps in them This may be CURA normal but with Slic3r I always get solid brims. I cannot see any settings in Cura to control the brim apart for the number of lines. Is this a normal CURA brim? Bed temp. was 60c; filament depth for 1st layer was 300 microns;1st layer speed was either 20 or 30m/s. Extruder temp for 1st layer was 220 for some 215 for the others (printing temp 210) All ideas appreciated

Hi has anyone had decent results with either of these products? Wondering about settings, especially bed temperature and material. I normally use hairspray on glass but suspect I might need to switch to blue tape for this. I failed miserably with their T-glass recently

I thought there were comments on this on our mega thread on under-extrusion ( I did say SOME users) but no problem I stand corrected I have not had problems with using 10 secs for minimum layer time and have found that better than 6 secs which I used to use, but I agree, I do often print multiple pieces.

Colorfabb no different from the UK in terms of delivery time - delivery cost pretty comparable too. Buy Faberdashery in 50m lengths and lay it on the floor

Interesting Illuminarti. I procrastinated for a while over retraction but in the end figured that given the design you would do more retracting than printing! Plus the UM2 seems to be having some problems in that area for some users. You may be right but I would not put a lot of money on the table on either suggestion! I have printed the standard test cube with holes for testing bridging using no retraction and have not suffered, which I think is because the travel distance is so small, like the chess piece. Normally I would contest that for this piece 7 seconds layer speed is probably too low but I have not run a small piece like this with the fans blasting out before so I guess you might get away with 7 seconds. I think I will give it a go today.

I think I am right in saying that Cura like most slicers only lets you differentiate between the 1st layer and the then all the other layers as one entity. Normally you want the first layer to be run at a slow speed , with a higher extruder and bed temperature and with a larger layer depth to ensure good adhesion of the first layer. In principle this is because the bed may not be completely flat and you need the first layer to absorb the unevenness and provide a flat surface for the 2nd and subsequent layers to sit on. Typical layer 1 settings would be Layer depth 300 microns Temperatures 5 or 10 degrees higher (although personally once I get to 210 and 60 I tend to use the same temp unless I am using a tricky filament). Speed 20m/s or 30 m/s No fan It is though less important for a small item such as a chess piece than for an item which has a large base such as most of the build platform. If you select the Advanced tab in Cura you will see initial layer thickness (i.e. layer 1) and bottom layer speed (i.e. layer1) where you can set alternate values to those you set under the Basic tab. I do not see anything in Cura for controlling the layer 1 temperatures so I guess (and may well be corrected) that if you want to have different temps for layer 1 you set them as the print and bed temps under the Basic tab and then change the temps manually via the printer controls once layer 1 has been printed. For that chess piece I suspect you want each printed layer to cool down as quickly as possible, especially up at the top part hence my suggested settings; whether you will get away without having any heat in the bed for layer 1 I am not sure. Possibly I guess, as the footprint of the chess piece is very small. One way to find out…J

Jonny is right, that is pretty difficult geometry for a beginner to print. You might want to first get this calibration set and see how you go printing these - http://www.thingiverse.com/thing:5573. But you probably do not want to wait- birthdays do not J I was thinking of printing this chess set. If I did I would start with… Layer height: no more than 100 microns, probably 50 if I can wait that long Speed: probably 15 m/s, might try 20m/s first Bed temperature for 1st layer 60c Extruder temp: 1st layer 210c 2nd layer onwards 200c (if you have a UM2 you may need to go higher than these) Fans: 100% on from layer 2 (it will sort of depend on the size but that is where I would start) Retraction: I think I would turn it off; actually I think I would definitely turn it off HTH

All I CAN say is that they are very helpful. I used them recently to buy some Taulman Nylon filaments. I had a particular question and they contacted Taulman in the States and got back to me the following day with a tele call.

So it is true then, eating Weetabix for breakfast does make you go blind

Used Faberdashery black for the first time last week and got good results at .2 layer 40 m/s 210c. Not tried other speeds/thicknesses yet. Colourfabbb are good too. NB above was not on a Multimaker 2

A bit off piste but I have a similar problem/requirement. I have started looking at barrelling machines. My wife is the main driver wanting to polish jewellery but there are clearly some out there that claim to smooth plastics too.

I have not spotted how to do this but is there a way to get Meshmixer to reduce the size of an object by, say, 25% (all axes)? I know you can manually rescale using alignment but you cannot, as far as I can see, use that to get an equal reduction on all 3 axes.,

This is a subject I do not understand; maybe someone can explain? I downloaded a hinged box recently, out of interest on how it would work. Going through the hinge was a bolt. Now in order to work the bolt cannot be fused to any part of the hinge – can you print that? I printed the box, tried to close it and it broke – unsurprisingly to me. Edit: OK it depends on hinge design; at least one part of the hinge cannot be fused to the bolt

Wash my glass with warm water or soapy warm water if I have been using nylon. No need to improve adhesion because I use hairspray. No time wasted trying to mix up solutions, no mess, just spray and go 20 seconds later. And it lasts for days and days.

Good subject to post Keith. Some years ago I used a 2-part epoxy for gluing resin based 1:43 car models and Colourfabb tech support confirmed it would be a good glue for PLA. In a rush though I got some Revell Contacta Professional from my local toyshop. It does not display the issue you noted but I cannot really say how strong the joints are; they seem OK but I have not yet attacked the assembly with a hammer